1. Field of the Invention
The present invention relates to an attenuating type phase shifting mask, and more particularly, to a pattern of attenuating type phase shifting mask and a method of manufacturing thereof.
2. Description of the Background Art
Recently, high integration and miniaturization of the semiconductor integrated circuit has been developed rapidly. Accordingly, fine processing of the circuit pattern formed on the semiconductor substrate has been developed rapidly. Particularly, photolithography has been recognized broadly as basic technology in formation of the pattern. Although various developments and improvements have been carried out so far for photolithography, miniaturization of the pattern is still on going and the need for improvement of the resolution of the pattern has been more and more increased.
Generally, the limit of resolution R (nm) in photolithography utilizing the demagnification exposure method can be represented as follows: EQU R=k.sub.1..lambda./(NA)
where .lambda. is a wavelength (nm) of light used, NA is the numerical aperture of a lens and k.sub.1 is a constant depending on resist process.
As can be seen from the above expression, values of k.sub.1 and .lambda. should be decreased and the value of NA should increased in order to improve the limit of resolution. In other words, the constant depending on resist process should be decreased, while the wavelength should be shortened and NA should be increased. However, it is difficult technically to improve the light source and lenses, and by shortening the wavelength and increasing NA, the depth of focus .delta. (.delta.=k.sub.2..lambda./(NA).sup.2) of light becomes shallower causing decrease of the resolution.
Referring to FIGS. 16(A), 16(B), and 16(C), a cross section of the mask, an electric field of exposure light on the mask and intensity of exposure light on a wafer when a conventional photomask is utilized will be described.
Referring to FIG. 16(A), a cross sectional structure of the photomask will be described. A metal mask pattern 20 made of chromium or the like is formed on a quartz glass substrate 10. Referring to FIG. 16(B), the electric field of exposure light on the photomask matches the mask pattern. As to the intensity of exposure light on the wafer, however, it is noted as shown in FIG. 16(C) that especially when the fine pattern is utilized, beams of the light transmitted through the mask intensity with each other at adjacent pattern images where those beams of light overlap with each other due to diffraction and interference of light. Consequently, the difference of the light intensity on the wafer is reduced resulting in decreased resolution.
In order to solve the above-mentioned problem, the phase shifting exposure method utilizing a phase shifting mask has been proposed in, for example, Japanese Patent Laying-Open Nos. 57-62052 and 58-173744.
Referring to FIGS. 17(A), 17(B) and 17(C), the phase shifting exposure method utilizing the phase shifting mask disclosed in Japanese Patent Laying-Open No. 58-173744 will be described.
FIG. 17(A) shows a cross section of the phase shifting mask. FIG. 17(B) shows the electric field of exposure light on the photomask. FIG. 17(C) shows the light intensity of exposure light on the wafer.
Referring to FIG. 17(A), the phase shifting mask includes a chromium mask pattern 20 formed on a glass substrate, and at every other aperture of mask pattern 20, a phase shifter 60 formed by a transparent insulating film such as a silicon oxide film is provided.
Referring to FIG. 17(B), as to the electric field of exposure light on the photomask provided by the light transmitted through the phase shifting mask, phases of the exposure light are alternately reversed by 180.degree.. In other words, beams of light are canceled with each other at adjacent pattern images where beams of light overlap with each other due to interference. Accordingly, as shown in FIG. 17(C), the resolution of pattern images can be improved, because of the sufficient difference of the intensity of exposure light on the wafer.
Although the above-mentioned phase shifting mask is very effective for a periodic pattern such as lines and spaces, the mask cannot be set appropriately in case of a complicated pattern because arrangement or the like of phase shifters is very difficult.
In order to solve the above-mentioned problem, an attenuating type phase shifting mask is disclosed in, for example, "JJAP Series 5 Proceedings of 1991 International Micro Process Conference, pp. 3-9" and in Japanese Patent Laying-Open No. 4-136854. The attenuating type phase shifting mask disclosed in the Japanese Patent Laying-Open No. 4-136854 will be described below.
FIG. 18(A) shows a cross section of the above-mentioned attenuating type phase shifting mask 500. FIG. 18(B) shows the electric field of exposure light on the mask. FIG. 18(C) shows the intensity of exposure light on the wafer.
Referring to FIG. 18(A), phase shifting mask 500 includes a phase shifting pattern 300 which is a predetermined exposure pattern, including a quartz substrate 10 through which the exposure light is transmitted, a transmitting portion 100 formed on a main surface of quartz substrate 10 for exposing the main surface of quartz substrate 10, and a phase shifter 200 for shifting the phase of the exposure light transmitting therethrough by 180.degree. relative to the phase of the exposure light transmitting through said transmitting portion 100.
Phase shifter 200 is an absorption type shifter film of a double-layered structure including a chromium layer 20 having 5-20% transmittance of exposure light, and a shifter layer 30 for converting the phase of the exposure light transmitting through transmitting portion 100 by 180.degree..
The transmittance of exposure light of phase shifter 200 is set to 5-25%, which is appropriate for lithography, since the thickness of the resist film after development is adjusted according to the transmittance as shown in FIG. 19.
The electric field on the mask of the exposure light transmitting through the phase shifting mask having the above-mentioned structure is as shown in FIG. 18(B). As to the intensity of the exposure light on the wafer, the light intensity at the edge of the exposure pattern is inevitably 0 as shown in FIG. 18(C), since the phase of the exposure light is reversed at the edge of the exposure pattern. Consequently, there is provided sufficient difference between the light intensity corresponding to the transmitting portion 100 and that corresponding to the phase shifter 200, so that the resolution can be improved.
The above-mentioned conventional technology, however, has the following disadvantages.
FIG. 20(A) shows positions of the attenuating type phase shifting mask placed in the exposure apparatus and a blind 70 for determining the exposure region of the exposure apparatus. FIG. 20(B) shows the electric field of the exposure light directly under the attenuating type phase shifting mask and blind 70. FIG. 20(C) shows the light intensity on the exposed material of the light transmitted through the attenuating type phase shifting mask and blind 70. FIG. 20(D) shows the region exposed by the light transmitted through the attenuating type phase shifting mask and blind 70.
Referring to FIG. 20(A), the region other than a chip pattern forming region (Lc) of the attenuating type shifting mask is covered with absorption type shifter film 20 in which pattern processing is not conducted. In a demagnification projection, exposure apparatus blind 70 which interrupts the light for determining the exposure region is provided at a prescribed position under the attenuating type phase shifting mask.
The aperture of blind 70 may have any width so long as it allows exposure of the chip pattern region, and therefore the width may be the same as the chip pattern region (Lc). However, since the position of blind 70 is controlled by the distance of about 1000 .mu.m (about 1 mm) and blind 70 is not positioned on the same plane of focus as the attenuating type phase shifting mask, the edge portion of blind 70 is not well focused out of focus. Thus, as shown in FIG. 20(A), the width of aperture of blind 70 (Lb) is set to about 1000 .mu.m wider than the chip pattern region (Lc) so that blind 70 does not overlap the chip pattern region (Lc).
Accordingly, in an ordinary photomask utilizing an interrupting film of chromium, for example, on the chip pattern, only 1/1000 of the light can be transmitted through chromium at most, so that the light passing through the gap between the chip pattern region and blind 70 will not expose the resist film on the semiconductor wafer.
In case of the attenuating type phase shifting mask, however, 5-20% of the exposure light passes through the gap between the chip pattern region and blind 70 as indicated by portion A of FIG. 20(B) because the transmittance of the absorption type shifter film serving as a material of the chip pattern is about 5-20%. Consequently, referring to FIG. 20(C), a region having the light intensity of I' which is 5-20% of the transmitted light I.sub.0 results between chip pattern region Lc and blind 70 as can be seen from the distribution of intensity of the light transmitted through the attenuating type phase shifting mask and through blind 70. Thus, referring to FIG. 20(D), there is formed a region 50 having the light intensity (I') and having the width of Ld of 5-20% around chip pattern region 30 (Lc.times.Lc).
When patterns of the attenuating type phase shifting mask are demagnified and transferred onto the semiconductor wafer utilizing the demagnification projection exposure apparatus having the above-described structure, exposure proceeds successively at a pitch of Lc which is the size of the chip pattern FIG. 21 shows the exposed regions on the semiconductor wafer when the wafer is exposed using the attenuating type phase shifting mask having the chip pattern of the size (Lc.times.Lc), utilizing the demagnification projection exposure apparatus.
In this case, since exposure proceeds at a pitch of Lc longitudinally and transversely, there is a region 50 having the light intensity (I') of 5-20% (I') as described above around the chip pattern provided by one exposure shot. This region 50 overlaps an adjacent region provided by another exposure shot. As exposure is repeated successively, region 50 is exposed overlapped with adjacent three regions 50 at each corner of the exposure region. As a result, the exposed region comes to include region 31 each of which is exposed with the light having the appropriate intensity I.sub.0 plus the intensity I', which is 5-20% of I.sub.0, and region 32 each of which is exposed with the light having the intensity I.sub.0 plus three times the intensity I'.
In regions 31 and 32 which are exposed in such overlapped manner, when a positive resist film, for example, is exposed, the thickness of the resist film decreases after development. On the other hand, when the absorption type shifter film having a high transmittance is used, the resist film is completely exposed so that the resist film is lost by development.